Networked environments such as the Internet allow people using local network devices to request information from remote network devices. Typically the local network device is a personal computer (PC) executing client software (e.g., a web browser); the remote network device is a server executing server software; and the network is the Internet. The requested information could be a web page containing information related to a hyperlink. For example, a user (web site visitor) may request a web page about baseball, from a more general web page about sports, by selecting a picture of a baseball with a pointing device such as a mouse. Once the selection is made, the local device sends a message to the remote device via the network. The message typically includes the domain and the name (i.e., the address) of the requested information, as well as information identifying the local device. Often, several messages are exchanged between the client and the server, further specifying the identity and capabilities of the local device.
Frequently, the proprietor of a particular web site desires information about the people requesting information from his web site. This demographic information allows the proprietor to tailor the content of his web site to the preferences of its visitors. For example, if a particular page is visited frequently by a particular demographic group (e.g., teen males), then the web site proprietor could target that demographic group when selecting content such as advertising banners. Ordinarily, only limited demographic information such as browser version and the address of the local device are available for collection by the server. This limited demographic information is typically inadequate for most content tailoring, because it contains very little personal information about the visitor.
A common technique to overcome this drawback is to develop a profile for each visitor by requesting or requiring each visitor to fill out a questionnaire on their first visit. Profile information gathered from the questionnaire is then stored locally in a small file called a cookie on the visitors PC. Alternatively, the information is stored on the server and correlated to an identification number or password associated with the visitor. Regardless of the manner in which the profiles are stored, each time the visitor returns to the web site, his profile information is associated with the requests he makes, creating transaction information for subsequent analysis and content tailoring.
However, this technique suffers from certain drawbacks. Often, visitors are irritated by the inconvenience of multiple questionnaires from multiple web sites asking the same questions over and over. In fact, this may have the detrimental affect of averting visitors from such sites. Further, many visitors are unwilling to disclose personal information for fear it may compromise their privacy. Often, visitors prefer anonymous browsing in an effort to reduce junk e-mail. Although many web sites guarantee anonymity, others do not. Investigating individual proprietors privacy policies while browsing can be burdensome. Still further, collection of demographic information by individual proprietors precludes analysis by that proprietor that goes beyond his particular web site. For example, a web site proprietor may want to know what percentage of visitors to any web site (not just his own) are teen males.